Multiple internal reflection (MIR) cells having a cylindrical MIR element (now called "circle cells" by some) appear to be initially disclosed in U.S. Pat. No. 3,017,802 issued on 1/23/62 to Walter Witt. U.S. Pat. No. 3,370,502, issued on 2/27/68 to Paul A. Wilks, Jr, described a cone-ended cylindrical MIR element. U.S. Pat. No. 4,595,833, issued on 6/17/86 to Donald W. Sting, disclosed ingenious optical elements for efficiently introducing infrared (IR) light into one end of a cone-ended cylindrical MIR element and for collecting IR light from the other end of such an element. The invention of Sting is commercially available from the Spectra-Tech division of Barnes Analytical, Stamford, Conn. as, for example, cell model 0005-001. Cylindrical MIR element cells are also commercially available from Wilks Scientific Corporation, South Norwalk, Conn. and from Axiom Analytical, Laguna Beach, Calif.
Cylindrical MIR element cells generally comprise a cell body. The cell body has a cavity within it. The cell body also has two apertures in it communicating with the cavity so that a cylindrical MIR element can be positioned in the cavity with the ends of the MIR element exposed to the exterior of the body at the apertures. An annular space remains in the cavity between the MIR element and the body. A fluid sample is flowed into this annular space. The MIR element is generally sealed to the body at each aperture in the body by the use of a single O-ring.
As discussed in the Witt patent, circle cells were developed for on-stream spectrometric analysis. As discussed in the Sting patent, cylindrical MIR element cells are uniquely applicable for on-stream spectrometric analysis. Despite the advancements made by Witt, Wilks, Sting and others over the years since 1962, however, a problem remained in on-stream analysis of industrial process streams using such cells. This problem was the possibility of leakage between the MIR element and one or both of the apertures in the cell body due to seal failure.
A leaking cell in an industrial on-line analyzer can be a serious problem, especially since such analyzers are usually not continuously attended. For example: (1) if the process stream is flammable, then there can be a fire or explosion hazard; or (2) if the process stream is toxic, then there can be a toxic release problem. It would be an advance in the art of on-stream chemical process analysis using cylindrical MIR element cells if a sealing means were developed that: (1) did not leak; or (perhaps more realistically) (2) that safely handled any leak that did occur.